Servier – Phlebolymphology

Endovascular treatment of venous thromboembolic disease: state of the art

Endovascular treatment of acute DVT
Anthony Comerota (USA)

The results of anticoagulation therapy for deep vein thrombosis are still unsatisfactory. After 5 years, there is a high level of valvular dysfunction (95%), venous hypertension (95%), obstructive iliac vein lesions (70%), calf muscle dysfunction (50%), and, finally, venous claudication and ulcers (30%). Direct venous thrombectomy can significantly improve the result of deep vein thrombosis treatment. It is characterized by significantly improved patency, lower venous pressure, less leg swelling, and fewer postthrombotic symptoms compared with anticoagulation (Level I data). The level of intramuscular pressures (a surrogate for venous pressure) decreased from 38 mm Hg to 12 mm Hg after direct venous thrombectomy. Therefore, if we have the contraindication to thrombolysis, we perform thrombectomy.

Another method that improved the treatment results is catheter thrombolysis, which improved health-related quality of life (71.4 vs 55.8; P<0.006) in patients with iliofemoral thrombosis. According to the results of randomized trial by Elsharawy et al (Eur J Vasc Endovasc Surg. 2002;24:209-214), the rate of patency was high after catheter thrombolysis vs anticoagulation (72% vs 12%; P<0.0001) and the rate of normal valve function was 89% vs 59% (P=0.041). According to the CaVenT study (Lancet. 2012;379:31-38), the development of postthrombotic syndrome with catheter thrombolysis was significantly better than anticoagulation therapy (iliofemoral thrombosis patency, 66% vs 47%; postthrombotic syndrome rate after 24 months, 41% vs 56%). Compared with catheter thrombolysis, pharmacomechanical thrombolysis can improve the result of treatment because this technology provides a higher rate of thrombolysis (92.3% vs 84.3%), requires less recombinant tissue plasminogen activator (33.5 vs 59.3), reduces the treatment time (24.2 vs 55.4), and results in more competent valves (57% vs 43%). The rate of residual thrombus is a significant predictor of deep vein thrombosis recurrence after all types of thrombolysis. So, in the group with a residual thrombus <50% the recurrent rate was 5%; it was 50% for the group with a residual thrombus rate >50%. In conclusion, the strategy of early thrombus removal is a validated and evidence-based method for treating acute deep vein thrombosis.

Endovascular treatment of obstructive DVT sequelae
Frederic Thony and Caroline Menez (France)

Endovascular treatment is the first-line treatment for postthrombotic syndrome. The fibroblastic invasion period occurs between 7 days and 1 month, and the period of recanalization, fibrosis, and retraction of thrombus occurs between 1 month and 6 months. The goals of treating an iliofemoral obstruction are to improve or heal the postthrombotic syndrome, to lower the risk of deep vein thrombosis recurrence, and to restore venous patency. In recent guidelines (AHA 2015, ESVS 2015) for severely symptomatic patients (or relevant postthrombotic syndrome) with iliac vein or vena cava occlusion, surgery or percutaneous endovascular recanalization maybe considered (class IIb, level C) for every patient who suffers from clinical symptoms related to venous sequelae and who would like improved or restored venous access.

Before treatment, clinical disability needs to be measured and a venous duplex examination needs to be performed with an estimation of the type of lesion (obstruction/synechia/compression), the extent of obstruction (popliteal-femoral evaluation), concurrent reflux, and cartography for treatment planning (venous access/landing zone). After a CT scan, Thony uses the grading of venous sequelae upstream of the common femoral vein. The three main veins of the thigh are graded from 0 to 2 (0, the vein is totally occluded; 2, large normal vein). According to this scale, he uses the three values to quantify venous sequelae as grade 0 (6 points; normal vein), grade 1 (4 to 5 points; minor sequelae), grade 2 (2 to 3 points; severe sequelae), and grade 3 (0 to 1 point; major sequelae). For the procedure, Thony prefers using the jugular vein access (80%), a 0.035 hydrophilic guide wire, and nitinol stents (with a high radial force for the common iliac vein and inferior vena cava and with high flexibility in the femoral vein). The patients receive aspirin (75 mg) for 1 month and anticoagulation treatment for 6 months. Thony also prefers using an intermittent pneumatic compression for the first night and compression stockings for 6 months.

A review of the literature shows a technical success rate of 97%, a major complication rate of 0% to 8.7%, a primary patency rate of 32% to 99%, a secondary patency of 66% to 96%. The treatment appears promising and safe; therefore, it should be considered as a treatment option while the evidence base is improved. According to the Thoney’s experience, the patency rate was high depending on values of venous sequelae. After a 24-month follow-up, the primary and secondary patency rates for the different grades of sequelae were as follows: grade 0, 87% to 100%; grade 1, 82% to 100%; grade 2, 58% to 90%; and grade 3, 8% to 63%. The author concluded that treating a deep vein thrombosis sequelae with an endovascular treatment is a highly efficient and long-lasting treatment. Patency correlated with the severity of the sequelae upstream of the common femoral vein, and concurrent deep or superficial venous insufficiency and extensive sequelae worsen the clinical results. The indication for endovenous treatment should be extended to every patient suffering from symptoms related to postthrombotic venous sequelae.

Vena cava filters
Patrick Mismetti (France)

According to the US Nationwide Inpatient Sample, between 1999 and 2010 in the US, there were 556 658 patients with a pulmonary embolism and vena cava filters were inserted in 17% of these patients. According to the Prospective European Registry from 2001 to 2012, among the 40 142 patients with a pulmonary embolism, only 1% received vena cava filters. According to the US National Register, there were no differences in the in-hospital mortality between patients with or without vena cava filters (mortality rate, 7.2% vs 7.9%). The use of vena cava filters in patients with a venous thromboembolism and no contraindications to anticoagulants increases the mortality rate (4.5% vs 3.5%) and significantly elevates deep vein thrombosis recurrence after 1 year (5.4% vs 3.7%). However, among the group of patients with unstable pulmonary embolism receiving fibrinolytic therapy, the mortality rate was lower after vena cava filter placement (7.6% vs 18%; P<0.001).

The PREPIC 1 study (N Engl J Med. 1998;338:409-415) demonstrated the negative effect of permanent vena cava filters for pulmonary embolism incidence (15.1% with vs 6.2% with no vena cava filter; HR, 0.37; 95% CI, 0.17-0.79) and for recurrent deep vein thrombosis (35.7% vs 27.5%; HR, 1.52; 95% CI, 1.02-2.27). The PREPIC 2 trial on the effects of retrievable vena cava filters showed that pulmonary embolisms at 3 months, pulmonary embolisms at 6 months, and death at 6 months were high, but not significant, in the group receiving vena cava filters. The conclusion was that there was no apparent benefit with the use of vena cava filters in stable pulmonary embolism patients without any contraindication to anticoagulants. However, the use of vena cava filters in venous thromboembolism patients with active bleeding (temporary or permanent contraindications to anticoagulants) significantly decreased mortality (9.5% vs 11.5%; P<0.003; HR, 0.68; 95% CI, 0.52-0.88).

In conclusion, vena cava filters are contraindicated in stable pulmonary embolism patients. For patients with an unstable pulmonary embolism, we can use vena cava filters with uncertain benefit; however, there is a clear benefit in the cases of recurrent venous thromboembolism despite adequate anticoagulation therapy and patients with bleeding and contraindications to anticoagulants. However, if the contraindications to anticoagulants are due to surgery, then vena cava filters are not beneficial. The use of retrievable vena cava filters needs an experienced multidisciplinary team to manage indications and retrieval.

Treatment of post-PE pulmonary hypertension
Helene Bouvaist (France)

Surgical pulmonary endarterectomy is the preferred treatment for chronic thromboembolic pulmonary hypertension, according to the 2015 ESC/ERS guidelines. However, 40% to 50% of such patients are inoperable due to distal lesions or comorbidities. In addition, surgical mortality is >10% if pulmonary vascular resistance is >15 WU, and residual chronic thromboembolic pulmonary hypertension occurs in 25% of patients. Balloon pulmonary angioplasty is a good method to treat chronic thromboembolic pulmonary hypertension. It is characterized by a decrease in mean pulmonary arterial pressure from 45.4 mm Hg to 24.0 mm Hg and a decrease in pulmonary vascular resistance from 942 to 327 dyne-sec/cm². Bouvaist performs angioplasty under local anesthesia using a femoral access, an international normalized ratio of 2 to 3, and a bolus injection of unfractionated heparin 1000- 2000 UI, and she uses a 0.014 hydrophilic wire and coronary/renal angioplasty balloon and 2 to 8 sessions for each patient with duration of 1.5 to 2 hours per session. The preliminary results (33 patients at 12 months) were good: the pulmonary vascular resistance decreased from 9 to 4.94 WU, mPAP decreased from 45 mm Hg to 34 mm Hg, and, out of 147 procedures, there were 6 minor hemoptysis and 2 pulmonary injuries. These results show that interventional balloon pulmonary angioplasty may be considered in patients who are technically inoperable or carry an unfavorable risk-benefit ratio for pulmonary endarterectomy (IIB, C). The ongoing RACE French multicenter (23 centers) study that is comparing the effects of riociguat balloon pulmonary angioplasty on pulmonary vascular resistance after 6 months for inoperable chronic thromboembolic pulmonary hypertension). The results of this trial will clarify the long-term prognostic effect of balloon pulmonary angioplasty.

Cancer-associated thrombosis: realities, actualities, and perspectives

Cancer and thrombosis: an intimate and dynamic relation
Anna Falanga (Italy)

The strong association between cancer and thrombosis was recognized for the first time in Paris by Armand Trousseau in 1865. Today, we know that approximately 20% of all venous thromboembolism cases occur in the population affected by cancer. Cancer increases the risk of thrombosis, and vice versa, thrombosis or hypercoagulation can affect cancer progression. Cancer-associated thrombosis has unique and multifactorial pathogenic mechanisms. Cancer tissues can activate blood clotting using three main routes: (i) expressing procoagulant factors in tumor cells, including tumor factor, tumor factor bearing microparticles, and heparinases; (ii) expressing adhesion receptors, which platelets and leucocytes can use to adhere to the endothelium; and (iii) releasing soluble mediators, such as inflammatory cytokines (IL-1β, TNF-α, VEGF) and proangiogenic factors (Best Pract Res Clin Haematol. 2009;22:49-60). Different tumor cell lines can activate coagulation by preferentially using one or more of these routes.

Tumors also differ in their capacity to express procoagulant activities (Haematologica. 2012;97:1173-1180). An important step in the last 15 years has been the recognition that tumor genetic programs drive the activation of clotting proteins so that oncogenes can upregulate the expression of procoagulant proteins in tumor cells. For example, patients with acute promyelocytic leukemia have a strong activation of blood coagulation with very dramatic thrombotic and bleeding manifestations. The mortality rate of this disease is mainly due to coagulopathy more than to leukemia per se, and it is recognized as one of the most curable forms of leukemia. In acute promyelocytic leukemia, the typical PML/RARα genetic lesion is associated with overexpression of anticoagulant activity (ie, tumor factor) and the occurrence of coagulopathy. During differentiation therapy with all-trans-retinoic acid (ATRA), which targets the molecular lesion, the bone marrow cell procoagulant activity expression decreases and the coagulopathy is simultaneously resolved (Semin Thromb Hemost. 2008;34:204-210).

Some coagulant activities are important for tumor biology and tumor growth and dissemination. Indeed tumor cells can attach to the endothelium by receptors and attract leucocytes and platelets, but they can also release cytokines, thus activating the endothelium and causing localized clot activation, microthrombi formation, and fibrin deposition. All of these mechanisms can help tumor cells extravasate to form a distant metastasis (Lancet Oncol. 2005;6:401-410). Tissue factor (TF) and TFbearing microparticles are the links between the blood coagulation cascade and cancer. The transmembrane glycoprotein, expressed by extravascular cells, is the main physiological activator of coagulation. TF is an important risk factor for venous thromboembolism in cancer patients, and its cellular properties play a pivotal role in primary tumor growth and metastasis in a broad set of cancer types (Semin Thromb Hemost. 2015;41:747-755). In addition, malignant cells produce microparticles (Blood. 2005;105:1734-1741) that can activate coagulation, and therefore, supports in vivo thrombosis (J Thromb Haemost. 2008;6:1517-1524; Am J Haematol. 2014;89:68-73).

The hypercoagulable state in cancer patients is a subclinical condition characterized by an alteration in thrombotic markers, without overt thrombosis. Clotting abnormalities increase at the start of chemotherapy and with tumor progression.

Measuring biological markers of thrombosis can have important clinical utility in predicting the occurrence of thrombosis, cancer survival, and cancer diagnosis. The occurrence of a venous thromboembolism in cancer patients may have important clinical implications, including an increase in treatment failure rates and quality of life deterioration in long-term cancer survivors. Therefore, identifying the risk factors for venous thromboembolism and establishing an effective preventive strategy are necessary to avoid life-threatening complications among cancer patients. The venous thromboembolism predictive value of circulating thrombotic biomarkers in cancer patients is under active investigation. Prospective studies are warranted to evaluate their clinical utility in identifying cancer patients who are at a high risk of thrombosis and who can benefit from thromboprophylaxis (Blood. 2013;122:2011-2018). A novel and promising approach for the stratification of patients according to their thrombosis risk include risk assessment models that have both clinical parameters and biomarkers.

Clinical prediction rules are appealing because they offer several potential benefits for practitioners, patients, and the health care system, such as a reduction in clinical uncertainty at the bedside and an improvement in the quality of life for patients. The most validated model is the Korana score, but scientists are trying to improve the quality of this score by adding biomarkers to clinical and standard laboratory parameters (Blood. 2010;116:5377-5382). For instance, adding D-dimer and soluble platelet selectin improves the score, which allows five levels of risk to be identified, thus distinguishing the intermediate risk patients better. The addition of ADAMTS-13 activity and F1+2 levels to Khorana or cancer-associated thrombosis scores considerably increases the predictive value of venous thromboembolism (Thromb Haemost. 2016;14:306-315). Therefore, the ASCO guidelines recommend that cancer patients be assessed for venous thromboembolism risk at the time of chemotherapy initiation and periodically after that. Individual risk factors, such as biomarkers or the cancer site, do not reliably identify cancer patients at high risk of venous thromboembolism. Risk assessment should be conducted based on a validated risk assessment tool, especially in the outpatient setting.

Prevention of CAT: a challenge for the personalized medicine
Grigorios Gerotziafas (France)

The major determinants of the risk of cancer-associated thrombosis include the type of cancer, the time from the diagnosis, the treatment, and the stage of the disease. In addition, venous thromboembolism risk factors not related to cancer, such as underlying comorbidities, hospitalization, and genetic variability polymorphisms, have to be considered in these patients. The incidence of symptomatic cancer-associated thrombosis differs in different cancer types and different stages of the disease. For instance, breast cancer, which is considered a cancer with a low-risk of venous thromboembolism, has almost the same level of risk as pancreatic cancer, which, in the metastatic stage of the disease, is known as a high-risk cancer. In addition to the increased risk according to the metastatic stage, the time from the cancer diagnosis is another important factor to be considered because the greatest number of the venous thromboembolism events occur in the first 3 to 6 months after diagnosis. A similar “critical interval” can be observed in lung cancer, which can also be stratified according to histological type, with adenocarcinoma being the most thrombogenic form and squamous cell carcinoma the least (Clin Oncol. 2009;21:425-426; J Thromb Haemost. 2008;6:601-608). When a venous thromboembolism event occurs, independently of both the type of cancer and the stage of the disease, it represents a bad prognostic factor for patient survival, which can be explained considering the type of cancer, which provokes venous thromboembolisms more aggressively, but also other factors are to be taken into account. Indeed, anticoagulation, the risk of bleeding, and the change in the general status of the patient can compromise the best chemotherapeutic or anticancer treatments.

Gerotziafas emphasized that the Khorana risk assessment model was constructed by a retrospective analysis of preexisting databases to evaluate only chemotherapyassociated thrombosis, it was restricted to ambulatory patients with specific cancer types, including stomach, pancreas, lung, gynecological, and testicular cancer, or lymphoma. It is not applicable to patients with breast cancer, the most frequent cancer in the community, and it can only be used before the initiation of chemotherapy (Blood. 2008;111:4902-4907; Cancer. 2007;110:1149-1161). In real life, the frequency of thromboprophylaxis in cancer patients is very low, and this lack of prophylaxis probably happens because oncologists are unaware of the risk of venous thromboembolism in their patients. The results of the COMPASS-CAT study, a prospective trial conducted to generate a new tool capable of detecting cancer and not just chemotherapy-associated thrombosis, were presented. Approximately 1300 patients with the most common cancers, such as breast, lung, colon, and ovarian, in different stages of disease, were included. Most of the patients were on treatment when they were assessed. Initially, data regarding comorbidities and risk factors for venous thromboembolism not related to cancer, such as family history of venous thromboembolism, underlying cardiovascular or lung diseases, and hospitalization, were introduced in the analysis. The number of events observed in this study was around 8%. This holistic approach allowed for the stratification of patients at a high level of risk (rate of venous thromboembolism during follow-up, 13.3%) or a moderate level of risk (rate of venous thromboembolism, 1.7%) and the selection of patients suitable for anticoagulant treatment.

Regarding biomarkers for hypercoagulability, it is almost impossible to find the best, single biomarker able to predict venous thromboembolism, and their dosage can only give a snapshot without any kinetic profile. In spite of these limitations, the weighted incorporation of several biomarkers in proposed risk assessment models for venous thromboembolism seems to improve their capacity to identify patients eligible for pharmacological thromboprophylaxis. In breast cancer patients, the levels and procoagulant activity of platelet-derived microparticles are interconnected with biological activity and the overall burden of cancer. Thrombin generation reflects the procoagulant properties of breast cancer and chemotherapy in the initial period of cancer diagnosis (BMC Cancer. 2014;14:991). In the ROADMAP study, an ongoing trial involving patients with lung adenocarcinoma, showed that hospitalization and time from cancer diagnosis were the most important clinical risk factors for cancerassociated thrombosis in patients with metastatic disease. In addition, baseline values of thrombin generation, heparinases, procoagulant phospholipids, and D-dimers were related to mortality and thrombosis. Therefore, when biomarkers are combined with clinical risk factors, they significantly improve the predictive value of the risk assessment models (Thromb Res. 2016;140[suppl 1]:S196). In conclusion, a contextualized approach to improve prophylaxis and treatment of venous thromboembolism needs to be elaborated by checking the patient’s profile, optimizing the patient management (especially in the critical intervals of increased risk), applying multidisciplinary approaches, and, finally, guaranteeing the quality of life, if not the life expectancy, of the patient.

EuroG20 on CAT, Bergamo 2016: challenges in the treatment of VTE
Ismail Elalamy (France)

Elalamy discussed the expert workshop on cancer-associated thrombosis, which was held in Bergamo, Italy. The aim of the meeting was to discuss unmet needs requiring clarification and produce Pragmatic and Practical Proposals for Patient care Promotion (5 Ps) and a European guideline for practices in complex cancer-associated thrombosis situations, for example, the limits and advantages of the Khorana score. The addition of biomarkers (eg, D-dimers and sP-selectin) to clinical and standard laboratory parameters can help predict the prediction of venous thromboembolisms and identify cancer patients at high or low risk for venous thromboembolisms (Blood. 2010;116:5377-5382). The Khorana and the Vienna scores need further validation to develop a more practical and specific score.

The meeting also discussed treatment strategy (dose and duration) in patients with tumors at a high bleeding risk, such as glioblastoma and/or cerebral metastasis. Glioblastomas are rare tumors that have a very high risk of thrombosis, and they are linked to a high risk of major bleeding (Thromb Res. 2015;136:1199-1203). Patients with glioblastomas receive slightly lower doses of antithrombotic medication than those with other cancers. Currently, there is no indication for thromboprophylaxis after the postoperative period in patients with brain tumors (Neuro Oncol. 2012;14[suppl 4]:iv73-iv80). The challenge now is to be able to propose thromboprophylaxis over a long period and to determine the right dose and duration. The optimization of prophylaxis in patients with brain tumors can probably be done by designing a double-blind, randomized controlled trial for the pharmacological prophylaxis of venous thromboembolism, identifying specific risk factors for bleeding and elaborating risk assessment models for bleeding.

Two further discussion points were to determine the occurrence of recurrent venous thromboembolism under low-molecular-weight heparin and define the risk of increasing dose and treatment by using a once or twice daily scheme. Indeed, despite effective treatment, a high risk of recurrence in cancer patients under anticoagulation has been observed in several studies (from 5% to 20%) (J Thromb Haemost. 2015;13:1010-1018). Once-daily treatment with low-molecular-weight heparin is as effective and safe as is twice-daily treatment with low-molecular-weight heparin in patients with no cancer (Cochrane Database Syst Rev. 2013;7:CD003074). Conversely, in cancer patients, the twice-daily scheme appears more effective without increasing the bleeding risk (J Oncol Pharm Pract. 2011;18:264-270). Currently, guidelines suggest shifting to a therapeutic dose of low-molecular-weight heparin if the recurrence occurs during treatment with direct oral anticoagulants or vitamin K antagonists for at least 1 month, increasing the dose by 25% if recurrence occurs during optimal treatment with low-molecular-weight heparin, and avoiding inferior vena cava filters, except for those with absolute contraindications to anticoagulation (J Thromb Thrombolysis. 2016;41:81-91).

For the antifactor Xa (anti-Xa) determination, the optimal therapeutic range at peak and trough has only been established for noncancer patients. The clinical relevance of monitoring seems to be low because there is no correlation between the level of anti-Xa and clinical efficacy or safety. The experts proposed that monitoring should be based on clinical context and reserved for special subgroups of cancer patients, such as patients with a low body weight, renal impairment, thrombotic recurrence, obese with a high glomerular filtration rate, high risk of bleeding, and fear of accumulation. Finally, regarding prophylaxis in myeloma patients, according to the International Myeloma Working group recommendations, aspirin still has a place in prophylaxis, but only for patients with ≤1 risk factor. However, low-molecular-weight heparin should be preferred depending on the type of myeloma treatment even in these patients. Patients with ≥1 risk factors should receive prophylactic low-molecularweight heparin.

Is there any place for DOAC in the prevention and treatment of CAT?
Isabelle Mahe (France)

The incidence of cancer and cancer-associated thrombosis is increasing; however, due to the progress in cancer-associated thrombosis treatments and supportive cares, a longer survival of patients with cancer-associated thrombosis has been observed. Direct oral anticoagulants–breakthrough molecules with a quick onset of action, a short half-life, few food and drug interactions, a broad therapeutic window at fixed doses, and no need for anticoagulant monitoring–have undoubtedly opened new scenarios for the management of venous thromboembolism, providing the opportunity to overcome vitamin K antagonist restrictions. However, to date, no specific clinical trial in cancer is available. We only have some data from registries and post hoc analyses of phase 3 clinical trials, which cannot be translated into a real-life setting. Furthermore, in the phase 3 trials, direct oral anticoagulants were compared with vitamin K antagonists and not with low-molecular-weight heparin, the gold-standard treatment for cancer-associated thrombosis, and few patients with low-risk cancer were included (5%), the definition of “active cancer” was heterogeneous among the studies, and little information on cancer characteristics and potential drug interference were provided (Thromb Res. 2015;136:582-589). Consequently, the guidelines do not recommend using direct oral anticoagulants as the first-choice therapy for the acute treatment of cancer-associated thrombosis.

Regarding extensive treatment, beyond the first 3 to 6 months, the comparator of the extended phase 3 trial assessing direct oral anticoagulants was variable because it was either vitamin K antagonists or placebo, and the subset of patients with active cancer was very small. The duration of treatment in cancer-associated thrombosis is still an open and debatable question. The DALTECAN study showed that the highest risk of major bleeding complications or cancer-associated thrombosis recurrence occurs in the first month of therapy. Despite this information, both recurrences and major bleeding persist over time with a similar rate, which is estimated to be around 10% at 1 year (J Thromb Haemost. 2015;13:1028-1035). For the management of cancerassociated thrombosis beyond the first 6 months of therapy, the guidelines suggest a more intensive treatment with low-molecular-weight heparin as the preferable option for higher risk patients and oral treatment discontinuation, and, for lower risk patients, oral anticoagulants (vitamin K antagonists or direct oral anticoagulants). For the prevention of cancer-associated thrombosis, there is currently no place for direct oral anticoagulants.

Direct oral anticoagulants seem to be an attractive option for patients with cancer associated thrombosis, but at this time, we have no data from studies focusing on patients with cancer-associated thrombosis. The ongoing trials comparing direct oral anticoagulants with low-molecular-weight heparin, such as the Hokusai-VTE cancer study and the Casta Diva study, should be able to answer these questions and determine if direct oral anticoagulants could represent a more convenient oral alternative that is at least as safe and effective as low-molecular-weight heparin, even in cancer patients.

Heparin centennial symposium

Heparin centennial: a century of clinical and scientific progress
Jawed Fareed (USA)

The year 2016 marked 100 years since the discovery of heparin. Heparin has had a very interesting life that has been full of difficulties. In 1916, heparin was placed in clinical trials, and regulatory bodies were put in place to regulate and develop guidelines for its use. In 1938, the Food and Drug Cosmetic Act for the Safety of Drug products was introduced; however, it was not until 1962 that the US Congress passed an amendment requiring premarketing proof of the effectiveness of drugs. Then, in 1966, there was another act put in place to regulate drug efficacy. Only in the 1970s was heparin labeling devised. The next 20 years was difficult for heparin due to its adverse effects, which led to it being withdrawn from the market. Today, the FDA has decided to reintroduce natural heparin, which is primarily due to the heparin contaminant crisis, which ultimately led to improvements. Guidelines are needed concerning natural, synthetic, and biotechnical heparin for multiple reasons, such as sourcing, required political controls, and other things, such as bovine issues. One study showed that the efficacy of heparin is the same regardless of its species of origin, especially for open-heart surgeries. Also, both bovine- and ovine-derived heparins have a higher level of activity when ingested. Today, heparin treatment is manageable, and, due to regulatory requirements, the risk of contaminants is no longer a concern. Dosing errors can now be minimized through a color-coding system and training; bleeding can now be easily reduced with an adequate dosage and monitoring; bovine spongiform encephalopathy (BSE) is no longer an issue because only BSE-free heparin is used and tested; and osteoporosis can now be avoided by using low-molecular-weight heparins. Thus, heparin is the only pleiotropic drug with multiple targets and broad applications. Furthermore, there are many innovations on the horizon, such as anticancer effects, neuromodulatory actions, and fertility and reproductive biology. The future success of heparin is dependent on increased awareness from the public and regulatory levels. Throughout its history, heparin has passed the test of the “survival of the fittest,” and it came out on top.

The pleiotropic effects of heparins: clinical relevance
Ludovic Drouet (France)

Glycosaminoglycans are long, nonbranching, negatively charged polysaccharide chains. Heparin is a highly sulfated glycosaminoglycan of natural origin. Only 20% of heparin chains are pentasaccharide sequences, which are responsible for its anticoagulant activity. Chains with pentasaccharide sequences lead to antifactor Xa activity, and, if the chains are long enough, they can additionally lead to antifactor II activity. At least 18 saccharide units are required to inactivate factor IIa; therefore, the antifactor IIa/antifactor Xa activity ratio only depends on the length of the molecule. The half-life of different molecules of heparin is inversely correlated with the length of the chains. Other than anticoagulant properties, including the chains with no anticoagulant activity, part of the complexity of the pharmacokinetics and pharmacodynamics of heparin can be explained by the interaction with the endothelial glycocalyx. Indeed, it has been demonstrated that heparin also has a broad antiinflammatory activity. Using the low-dose anticoagulant 2-O,3-O-desulfated heparin (ODSH), it was shown that most of the anti-inflammatory pharmacology of heparin was unrelated to its anticoagulant activity. In mice, ODSH was more effective than heparin in reducing selectin-mediated lung metastasis from melanoma, and it inhibited the airway inflammation mediated by the receptor for advanced glycation end product from the intratracheal high mobility group box protein-1 (Am J Physiol Cell Physiol. 2010;299:C97-C110).

Furthermore, inflammatory conditions provoke the generation of extracellular DNA traps, which are recently discovered large DNA fibers produced by neutrophils (neutrophil extracellular traps [NETs]). The recently described mechanism, called NETosis, implies that leukocytes can release a meshwork of chromosomal DNA, which includes histones and granular antimicrobial proteins, such as myeloperoxidase or neutrophil elastase, to trap and kill microorganisms. Extracellular DNA traps have been linked to several diseases, including sepsis and venous thrombosis. Unfractionated and low-molecular-weight heparins can block the DNA–von Willebrand factor (vWF) interaction, which contributes to thrombus initiation and progression (Arterioscler Thromb Vasc Biol. 2014;34:1382-1389). Such mechanisms, independently of the anticoagulant properties of heparin, have a great clinical relevance because they may contribute to the observed beneficial effects of heparin in the treatment of sepsis patients. It has been shown that nonanticoagulant heparin, purified from clinical grade heparin, binds histones, prevents histone-mediated cytotoxicity in vitro, and reduces mortality from sterile inflammation and sepsis in mouse models without increasing the risk of bleeding (Blood. 2014;123:1098-1101). In cancer patients, thrombosis is also related to the NETosis system. It has been demonstrated that neutrophils isolated from patients with gastric cancer presented a significantly enhanced NET formation compared with those in healthy controls, which can be prevented by heparin (Int J Clin Exp Pathol. 2015;8:14075-14086; Thromb Res. 2016;139:56-64). In conclusion, all the pleiotropic activities of heparin are extremely important and can have great clinical implications.

From unfractionated heparins to low molecular weight heparins and the evolution of synthetic heparins
Jeanine Walenga (USA)
In terms of anticoagulants, only 20% of heparin chains are now familiar and in use today, which leads to an enormous potential for these drugs. Throughout the history of heparin, many important discoveries have been made. In the 1930s, with the first-generation heparins, unfractionated heparin originated from bovine and porcine sources. From the 1970s to 2000, the second-generation heparins–low molecular- weight heparin and ultra-low-molecular-weight heparin–were being used clinically. From 2000 to 2015, the third-generation heparins–pentasaccharides, the first synthetic heparin–started being used. Currently, these pentasaccharides are now being made using different bioengineering processes, such as chemo-enzymatic synthesis, to make the production more cost-effective, and it allows the other 80% of chains (the nonanticoagulant components of heparin) to be studied for additional clinical uses. Unfractionated heparin is still used, and it will continue to be used for a variety of clinical situations; however, low-molecular-weight heparin led to important changes in the scope of how we treat thrombosis.

Heparin fractionation has resulted in the isolation of high-, medium-, and low molecular- weight components that have defined biological activities. Low-molecular weight heparin was the focus of drug development, and, to this day, low-molecular weight heparins continue to be the standard of care for prophylaxis and treatment of venous thrombotic disorders. Additional fractionation methods for heparin, including ion exchange chromatography and affinity chromatography, have also provided components with differential pharmacologic profiles. With the advent of the generic low-molecular-weight heparin enoxaparin, it became important to determine how the low-molecular-weight heparins compare to each other, and to determine if they have some differences in molecular weight, structural chains, and chemistry.

Since the discovery of the antithrombin effect of the heparin-binding protein in the 1970s, structural activity studies have been ongoing. This work then led to the isolation of the high antithrombin affinity heparin fractions. The most active fractions were found to contain a unique region containing antithrombin binding consensus sequences. In the late 1970s, the Choay group (Paris) isolated octasaccharides and hexasaccharides with a high affinity for antithrombin from the lower-molecular-weight fractions of heparin, and then they discovered the pentasaccharides. In 1968, the heparin cofactor antithrombin was discovered, then, in the late 1970s, data was ascertained about heparin binding to antithrombin; thereby, inhibiting factor Xa, which inhibits thrombin activity within the coagulation cascade; therefore, inhibiting thrombin production. Later, it was shown that pure antifactor Xa activity had antithrombotic activities. Throughout the 1980s, studies carried out with pentasaccharides led to it becoming a clinical drug. Due to the success of these pentasaccharides, there is more interest today in synthetic heparins to reduce the heparin-induced side effects while maintaining the antifactor Xa activity. Heparins are polytherapeutic, and they can be used for a wide range of situations beyond their conventional use as anticoagulants. Synthetic- and biotechnology-based heparin-related agents will emerge for certain clinical indications and new therapeutic targets, but natural source heparins will remain the agents of choice for broad clinical use.

Heparins: from pharmacokinetic properties to clinical practice
Patrick Mismetti (France)

The pharmacokinetic properties of low-molecular-weight heparin allowed us to move beyond the limits of unfractionated heparin and consider these drugs as an option in several clinical conditions. Indeed, the higher specificity and reduced variability, which is due to low binding to plasma and matrix proteins, resulted in fixed-dose usage without monitoring. Furthermore, the longer elimination half-life, which is due to an almost exclusive renal excretion, resulted in once-daily doses and long-term treatments. Meta-analyses have shown that low-molecular-weight heparin is more effective than is unfractionated heparin, which significantly reduces the occurrence of major bleeding during the initial treatment of venous thromboembolisms (Cochrane Database Syst Rev. 2004;18:CD001100). In addition, low-molecular-weight heparin appears significantly superior to vitamin K antagonists regarding the risk reduction in recurrent venous thromboembolism in cancer patients, with comparable safety to vitamin K antagonists (Thromb Res. 2015;136:582-589). Population pharmacokinetic studies, using antifactor Xa activity measurements, have shown that patient characteristics may influence the residual pharmacokinetic variability of low-molecular-weight heparin, with age, body weight, and renal function being significant covariates (Br J Clin Pharmacol. 2003;56:96-103; Clin Pharmacol Ther. 2005;77:542-552). Interestingly, maintaining the same doses of low-molecular weight heparin throughout pregnancy resulted in a progressive reduction in mean and peak antifactor Xa activities. Therefore, the administration of doses normalized for body weight variations to counteract pharmacokinetic changes during different stages of pregnancy is recommended. However, the use of a standard dosing regimen in patients with moderate or severe renal impairment resulted in elevated levels of antifactor Xa and reduced low-molecular-weight heparin clearance (Ann Intern Med. 2006;144:673-684). Thus, dose adjustment or laboratory monitoring of lowmolecular- weight heparin is commonly recommended in these patients to reduce the risk of major bleeding. In conclusion, the benefit-risk ratio of these molecules can be improved by consistently taking into consideration their pharmacokinetic properties.

Glycosaminoglycans and beyond
Job Harenberg (Germany)

As of September 2016, 181 clinical studies on low-molecular-weight heparin were registered on clinicaltrials.gov, with particular reference to rare localization of thrombosis, acute gastrointestinal bleeding, renal failure and hemodialysis, infectious diseases (pancreatitis), intraocular-infusion cataract operation in children, cardiac indications (coronary artery bypass surgery, aspirin discontinuation, bridging), superficial vein thrombosis (dose finding), and other specific indications (in vitro fertilization). In the future, other than improving the current use of heparin as a therapy for traditional circumstances and more specific indications, the nonanticoagulant activities should be considered just as important. Most probably, following the present scenario, attention should be focused on the pleiotropic effects of glycosaminoglycans (antitumor, anti-inflammatory, antiproliferative, anti–Parkinson) on the inhibitors of heparinases and heparin-derived polysaccharides without antithrombin binding sites, which are currently under development for the treatment of nonthrombotic diseases (Graefes Arch Clin Exp Ophthalmol. 2015;253:829-837; PLoS One. 2015;10:e0118798)

The development of low molecular weight heparins and their impact on VTE
Sylvia Haas (Germany)

Heparin is one of the oldest drugs still in widespread use. Heparin and vitamin K antagonists have been the main anticoagulant drugs for more than 70 years. Although heparin was first discovered a century ago, many years passed before it was mass-produced and used as an anticoagulant. Today, the main challenges of venous thromboembolism treatment remain almost the same–to prevent acute pulmonary embolism, thrombus extension, venous thromboembolism recurrence, and long-term sequelae and facilitate thrombus regression. Low-molecular-weight heparins have been the cornerstone of venous thromboembolism treatment for several decades. Despite the development of a monotherapy treatment with two oral factor Xa inhibitors, low-molecular-weight heparins will remain a key element in venous thromboembolism treatment. Low-molecular-weight heparins cannot be replaced in the near future in pregnant patients and other patient subgroups.

VTE prevention with heparin and other glycosaminoglycans in major orthopedic surgery: a personal journey
Graham Turpie (Canada)

Since 1962, venous thromboembolism was recognized as an important factor in orthopedics and a major cause of death in the wards, particularly in young individuals. Six years later, when the first hip replacement was done, Sir John Charnley recognized that “thromboembolic complications are the commonest of post-operative complications following hip surgery and the single largest cause of death.” These findings highlighted the relevance of venous thromboembolism in orthopedics, and therefore, the importance of prevention. Orthopedic surgery involving patients who are at risk for thrombosis and bleeding is a perfect setting to study new anticoagulants. The first randomized controlled trial using enoxaparin to prevent deep-vein thrombosis in patients undergoing elective hip surgery spawned the developments of new anticoagulant prophylaxis with low-molecular-weight heparin (N Engl J Med. 1986;315:925-929). In double-blind trials using the factor X inhibitor in major orthopedic surgery, fondaparinux showed a major benefit over enoxaparin, achieving an overall venous thromboembolism risk reduction >50% without increasing the risk of clinically relevant bleeding. These studies revealed that fractionation of heparin resulted in greater advantages for the patients (Arch Intern Med. 2002;162:1833-1840). Finally, in 2011, the initial evaluation of direct oral anticoagulant drugs through four randomized controlled trials looking at rivaroxaban compared with enoxaparin in an orthopedic setting showed an approximate 50% reduction in the risk of venous thromboembolism with an oral agent compared with enoxaparin, and it did not increase the risk of major bleeding complications (Thromb Haemost. 2011;105:444-453). However, phase 3 clinical trials are not the final word, meaning that real-world evidence, particularly from phase 4 noninterventional studies, is needed to evaluate the safety and effectiveness of new treatments further in clinical care settings. In this regard, the XAMOS study, which enrolled 17 701 unselected patients from 252 centers in 37 countries, confirmed the favorable benefit-risk profile of rivaroxaban that was seen in the RECORD program and routine clinical practice (Thromb Haemost. 2014;111:94-102). Thus, in 2016, the journey continues with direct oral anticoagulants.

Stepwise evolution of heparins and pentasaccharide: the French logic
Pierre Willaime (France)

At the end of the 1950s, heparin was thought to be a lifesaving drug to be used with caution because of the pathogenic effects. At the same time, open-heart surgery became almost standard. In addition, this combination led to the necessary and continuous research on this drug. Initially, Choay became interested in the drug because of the necessity of using heparin prophylactically, to administer small volumes using an easy to inject system, and to find a way to keep it sterile. These requirements led to multicenter studies. Facing two very different approaches, in the late 1970s, two studies were undertaken: the first to try to fractionate heparin and isolate the antithrombin factor, and the second, once fractioned, to produce it synthetically. In 1985, the second study resulted in a hexasaccaride becoming the first low-molecular weight heparin on the international market. Studies on hexasaccaride continued until the late 1990s when Sanofi decided to put the project on hold as the synthesis process was becoming too costly; therefore, making the product commercially nonviable. Two years later, the process was shortcut, and it was once again commercially feasible; the product was launched on the market as a pentasaccharide. The world market for heparin is worth 8 billion dollars, 85% of which is due to low-molecular-weight heparin, it has a growth rate of about 6%, and the number of studies today on heparin is in the hundreds; therefore, heparin can be compared to a “distinguished, attractive old lady who maintains and offers appealing features.”

The role of heparins and other glycosaminoglycans in VTE management: recent international guidelines
Andrew Nicolaides (Cyprus)

Nicolaides discussed the recent international guidelines produced by the International Faculty, which meets every 4 to 5 years to revise and update the document. The aim of these guidelines is to provide a clear and concise account of the evidence regarding efficacy or harm for various methods available to prevent and manage venous thromboembolisms. The evidence is presented for several outcomes, such as symptomatic deep vein thrombosis or pulmonary embolism, fatal pulmonary embolism, mortality, postthrombotic syndrome, and, interestingly, even for asymptomatic deep vein thrombosis. The decision to use asymptomatic deep vein thrombosis as well as symptomatic deep vein thrombosis and pulmonary embolism was based on several reasons:

• There is a relationship between symptomatic and asymptomatic venous thromboembolism, and a reduction in the incidence of asymptomatic deep vein thrombosis has been correlated with a reduction in symptomatic deep vein thrombosis and pulmonary embolisms.
• Reducing silent deep vein thrombosis is related to a reduction in clinical deep vein thrombosis, clinical pulmonary embolisms, and fatal pulmonary embolisms.
• Regulatory authorities have recognized asymptomatic proximal deep vein thrombosis as a valid end point for clinical trials in drug evaluation.
• Relatively few pulmonary embolisms occur in patients with symptomatic deep vein thrombosis, the majority of pulmonary embolisms and fatal pulmonary embolisms occur in patients with asymptomatic deep vein thrombosis.

Therefore, it has been argued that asymptomatic deep vein thrombosis is an important stage of venous thromboembolism that has not yet manifested itself and it cannot be ignored. Several clinical trials have compared unfractionated heparin or lowmolecular- weight heparin first with placebo and later, for ethical concerns, with active treatment.

The level of evidence for low-molecular-weight heparin and intermittent pneumatic compression has to be considered high for both moderate- and high-risk patients undergoing general or orthopedic surgery. Low-molecular-weight heparin is considered the method of choice for general surgery. Fondaparinux has a moderate level of evidence for general surgery and a high level for elective orthopedic surgery, showing it to be the most effective treatment available in elective orthopedic surgery. In addition, if low-molecular-weight heparin is used, it is recommended to start the treatment before or after surgery. Conversely, if fondaparinux is used, the prophylaxis should begin at least 6 hours after surgery to minimize the bleeding risk. Regarding treatment duration, extended prophylaxis (4 to 6 weeks) is recommended for total hip replacement patients (based on the Hull study results), for cancer surgery (based on the Enoxacan study), and for major abdominal surgery (Rasmussen study). In addition, the Exclaim multicenter trial, in spite of a great reduction in the incidence of venous thromboembolism, showed that, in medical patients, the price to be paid is an increased risk of bleeding.

VTE treatment with heparin and other glycosaminoglycans
Russell Hull (Canada)

The Achilles heel of unfractionated heparin was monitoring and dosage. These concerns led pharmaceutical and medical research to make an effort to tweak the drug to overcome these limitations. Low-molecular-weight heparins turned out to be more effective, safer, with a lower risk of thrombocytopenia, and, ultimately, had enormous benefits for patients, shifting venous thromboembolism treatment from the hospital setting to home care. Currently, low-molecular-weight heparins remain the gold-standard therapy for cancer-associated thrombosis in both acute and long-term phases of therapy. In addition to being superior to unfractionated heparin, low molecular- weight heparin produces a better recanalization than standard treatment with vitamin K antagonists.

The rationale for the management of VTE in cancer with heparins: current and future perspectives
Lord Professor Ajay Kumar Kakkar (UK)

Kakkar covered four areas: (i) the challenge of managing venous thromboembolism in cancer patients; (ii) the rationale for using low-molecular-weight heparin as the gold standard treatment for cancer-associated thrombosis; (iii) the duration of secondary prevention; and (iv) the value of these strategies. Cancer patients, despite long-term anticoagulation, have a three-fold higher risk of recurrent venous thromboembolism and a two-fold higher risk of major bleeding complications when compared with noncancer patients (Blood. 2002;100:3484-3488). In this population, strategies that can optimize the treatment by reducing the risk of recurrence and bleeding to improve the quality of life are indispensable.

Findings from several randomized controlled trials showed that low-molecular-weight heparins are more effective and safer than oral anticoagulant therapy in preventing recurrent thromboembolism in patients with cancer-associated thrombosis. In the CLOT study, 676 patients with cancer-associated thrombosis were randomly assigned to receive dalteparin once daily for 5 to 7 days and a coumarin derivative for 6 months or dalteparin alone. During the 6-month study period, 27 of the 336 patients in the dalteparin group had recurrent venous thromboembolism vs 53 of 336 patients in the oral anticoagulant group (hazard ratio, 0.48; P=0.002). No significant difference between the dalteparin group and the oral anticoagulant group was detected in the rate of major bleeding (N Engl J Med. 2003;349:e23). Similar results were observed in the CATCH trial with the use of tinzaparin (JAMA. 2015;314:677-686). Furthermore, the guidelines recommend low-molecular-weight heparin for at least 3 to 6 months and possibly indefinitely for patients with an active malignancy.

However, there is little data supporting treatment with low-molecular-weight heparin beyond 6 months. The DALTECAN study was conducted to determine the safety of dalteparin between 6 and 12 months for cancer-associated venous thromboembolism. Of the 334 patients enrolled, 185 and 109 completed 6 and 12 months of therapy, respectively; 49.1% had deep vein thrombosis; 38.9% had pulmonary embolism; and 12.0% had both on presentation. The overall frequency of major bleeding was 10.2%. Major bleeding occurred in 3.6% of the patients in the first month, and 1.1% and 0.7% per patient-month during months 2 to 6 and 7 to 12, respectively. Recurrent venous thromboembolism occurred in 11.1%; the incidence rate was 5.7% for month 1, 3.4% during months 2 to 6, and 4.1% during months 7 to 12. Thus, major bleeding was less frequent during dalteparin therapy beyond 6 months. The risk of developing major bleeding complications or venous thromboembolism recurrence was greatest in the first month of therapy and lower over the subsequent 11 months (J Thromb Haemost. 2015;13:1028-1035). There are still a number of outstanding clinical questions to address. How do we determine those patients who should have extended low-molecular-weight heparin therapy? How does the natural history of venous thromboembolism affect that of cancer? How will changes in cancer therapy affect the burden of venous thromboembolism? More research is certainly required in those areas. Moreover, the challenge of contemporary health care should be to demonstrate the value of aggressive strategies to prevent recurrent venous thromboembolism and improve outcomes because venous thromboembolisms have a terrible impact on cancer outcomes.

Current consensus on the management of VTE: a focus on heparin
Samuel Goldhaber (USA)

The 2016 American Heart Association statistics show that pulmonary embolism is still the third cause of cardiovascular death in the US, and the prevalence of venous thromboembolisms is increasing. Anticoagulation is the foundation of venous thromboembolism treatment. Parenteral anticoagulants for the initial management of venous thromboembolism include unfractionated heparin, low-molecular weight heparin, fondaparinux, and direct thrombin inhibitors (eg, argatroban and bivalirudin). At this time, unfractionated heparin is used with advanced therapy, such as thrombolysis, embolectomy, or inferior vena cava filters. Low-molecular-weight heparin or fondaparinux are reserved for patients who only require anticoagulation, or they are used for 5 days before switching to dabigatran or edoxaban. Direct thrombin inhibitors are the choice treatment for confirmed or suspected heparin induced thrombocytopenia. The guidelines from the American College of Chest Physicians (ACCP) for venous thromboembolism treatment, which are based on less bleeding with direct oral anticoagulants and greater convenience for patients and health care providers, now suggest that a direct oral anticoagulant should be used in preference to a vitamin K antagonist for the initial and long-term treatment of venous thromboembolism in patients without cancer (Chest. 2016;149:315-352).

Currently, three anticoagulation strategies are available: (i) classic strategy, which involves the overlap of low-molecular-weight heparin and unfractionated heparin with warfarin; (ii) switching strategy, which uses primarily low-molecular-weight heparin for the first 5 days and then switches to dabigatran or edoxaban; and (iii) an oral monotherapy regimen strategy, which uses rivaroxaban or apixaban as a single-drug approach (Lancet. 2012;379:1835-1846). Rivaroxaban, apixaban, and edoxaban work directly against factor Xa, and they are similar to the oral fondaparinux. Conversely, dabigatran is a direct thrombin inhibitor that is similar to argatroban or bivalirudin. For cancer patients, the updated ACCP guidelines suggest beginning treatment with 3 months of low-molecular-weight heparin as a monotherapy, and then either continuing low-molecular-weight heparin or switching to a different anticoagulant as long as the cancer remains active.

We are currently awaiting the results of ongoing trials comparing direct oral anticoagulants with low-molecular-weight heparin. The biggest change in recent years concerns the concept of managing venous thromboembolism mostly as a chronic inflammatory illness, not a “one-shot” event that is “cured” with 3 to 6 months of anticoagulation. Indeed, an extended duration of anticoagulation is often needed, and, as soon as the extended duration of anticoagulation is discontinued, the rate of new pulmonary embolisms or a new deep vein thrombosis is likely to increase (Haematologica. 2007;92:199-205). The anti-inflammatory effects of heparins and glycosaminoglycans can help to prevent recurrent venous thromboembolism.

Superficial venous thrombosis

Introduction
Claudio Allegra (Italy)

Many questions remain in the treatment of superficial vein thrombosis, including the role of direct oral anticoagulants on superficial vein thrombosis; the real activity of these non–vitamin K antagonist oral anticoagulants on the recanalization of the venous thrombosis; the fibrinolytic action of rivaroxaban; the influence of rivaroxaban on the rate of recanalization; and the role of heparin in these processes?

Epidemiology and the clinical picture
Marie-Antoinette Sevestre-Pietri (France)

Between 1998 and 2003, the frequency of deep vein thrombosis decreased from 1.83 to 1.57 per 1000 and the incidence of pulmonary embolisms increased from 0.6 to 0.81 per 1000; however, there is no direct data on superficial vein thrombosis (Thromb Haemost. 2016;116(5):967-974). The STEPH community-based study, which was conducted on 265 687 inhabitants of Saint Etienne, France, showed that the annual diagnosis rate of superficial vein thrombosis (0.64%) was lower than the rate of deep vein thrombosis (1.24%) and similar to that of pulmonary embolisms (0.6%) (J Thromb Haemost. 2014;12(6):831-838). The risk factors for superficial venous thrombosis include varicose veins, hormone replacement therapy, thrombophilia, cardiac-respiratory failure, compression therapy, and a history of pulmonary embolism and deep vein thrombosis. According to the OPTIMEV study, the risk factors for deep vein thrombosis/pulmonary embolism in the case of superficial vein thrombosis are age >75, active cancer, inpatient status, superficial vein thrombosis on a non-varicose vein (Thromb Haemost. 2011;105:31-39). Superficial venous thrombosis affects the great saphenous trunk in 53% to 68% (33% above the knee and 15% to the saphenofemoral junction), the small saphenous trunk in 13% to 15% (saphenopopliteal junction 33%), and other veins in 59% to 63% of cases. In 4% to 12% of cases, the thrombus extends to the perforating vein. Superficial vein thrombosis has clear clinical signs, but clinical examinations cannot diagnose associated deep vein thrombosis; therefore, an ultrasound analysis is necessary because the combination of deep vein thrombosis and superficial vein thrombosis is a frequent occurrence (up to 25%).

Instrumental evaluation
Pier-Luigi Antignani (Italy)

Superficial vein thrombosis on a healthy vein may be associated with a systemic disease, and there may be another reason for the thrombosis, such as a malignancy elsewhere, an autoimmune disease, Berger’s disease, or an inherited blood clotting disorder. Migratory thrombophlebitis requires a more detailed evaluation of the patient to find the malignant lesion (eg, CT scans, mammography, colonoscopy, serum carcinoembryonic antigen, and prostate-specific antigen). The duplex ultrasound is still the diagnostic method of choice for finding a vein thrombosis, and, during the examination, it is important to verify whether the clot is adherent to the vein wall, evaluate the deep venous system, and consider the thrombosis as a deep vein thrombosis if the thrombus is located 2 cm from saphenofemoral junction or the saphenopopliteal junction. Venography is rarely required to diagnose a superficial vein thrombosis. If information on the pelvic veins or iliac vein outflow tract is required, CT venography, if available, is the preferred method. The consensus document for the evaluation of superficial vein thrombosis was published in 2012 (Int Angiol. 2012;31(3):203-216.

What we learned from clinical trials
Isabella Quere (France)

An analysis of the STEPH, POST, and OPTIMEV epidemiological studies showed that the frequency of deep vein thrombosis and pulmonary embolism is 18.1% and 6.9%, respectively. The predictors of symptomatic vein thromboembolism (deep vein thrombosis/pulmonary embolism) and extension/recurrence of symptomatic pulmonary embolism include male sex, previous deep vein thrombosis or pulmonary embolism, cancer, and a superficial vein thrombosis in a non-varicose vein. Symptomatic extensions are common complications of superficial vein thrombosis, and they are associated with a significant risk of venous thromboembolic complications, irrespective of whether or not they reach the saphenofemoral junction. The STENOX and STAFLUX trials showed that 12 days or 30 days of heparin treatment (intermediate dose of low-molecular-weight heparin or unfractionated heparin), respectively, is not long enough, and there is a rebound effect. The CALISTO trial showed that patients with superficial vein thrombosis who were treated with fondaparinux had a significant reduction in symptomatic venous thromboembolism, superficial vein thrombosis extension, and superficial vein thrombosis recurrence, and there was no rebound effect. The 2012 American College of Chest Physicians guidelines recommend using a prophylactic dose of fondaparinux or low-molecular-weight heparin for 45 days in patients with a superficial vein thrombosis at least 5 cm in length (2B), and it is recommended to use fondaparinux 2.5 mg daily over a prophylactic dose of low molecular- weight heparin.

Heparin: when and why?
Eva Kalodiki (UK)

Heparin is a pleiotropic drug that has anticoagulant and anti-inflammatory effects. According to the 2012 Cochrane review, low-molecular-weight heparin appears to reduce the extension or recurrence of superficial venous thrombosis or both vs placebo, whereas the available data did not show any significant effect on venous thromboembolism. The new classification for the management of superficial vein thrombosis–the SEAP classification (S [I, isolated; M, multifocal; D, deep vein thrombosis]; E [N, non-varicose vein, V, varicose vein, R, reticular]; A [A, above knee; B, below knee; S, saphenous; J, junctional]; and P [S, spontaneous; M, malignancy; I, inflammatory; T, trauma; H, hematological])–was presented. A registry that is based on the SEAP classification is needed for the management of superficial vein thrombosis that may identify the categories that would benefit from heparin or other treatment.

Is there a place for NOACs?
Jawed Fareed (USA) and Eduardo Ramacciotti

Although direct oral inhibition is likely efficacious and safe, no studies have been completed on its use for patients with superficial venous thrombosis; therefore, the efficacy, safety, appropriate dose regimens, and treatment duration are not known. The SURPRISE trial, an ongoing, prospective, randomized, open-label, blinded, adjudication trial, is evaluating the efficacy and safety of rivaroxaban 10 mg once daily vs fondaparinux 2.5 mg once daily for the treatment of superficial venous thrombosis in high-risk patients over a period of 45 days. The aim of SURPRISE is to demonstrate noninferiority of rivaroxaban vs fondaparinux in the prevention of the combined efficacy end point of thrombus progression, superficial venous thrombosis recurrence, deep vein thrombosis, pulmonary embolism, and death. In addition to the SURPRISE trial, the multicenter randomized controlled trial RASET is evaluating the effects of rivaroxaban vs placebo in patients with superficial venous thrombosis.

Venous thromboembolic diseases
Marie-Antoinette Sevestre-Pietri (France)

The risk of venous thromboembolism is lower in men than in women; however, in women, venous thromboembolisms are related to hormonal conditions, such as pregnancy, contraception, ovulation-inducing hormones, and hormone replacement therapy. In addition, thrombophilia is often identified during pregnancy, and 40% to 50% of women who have a venous thromboembolism during pregnancy have thrombophilia. Different combined oral contraceptives have different impacts on the risk of venous thromboembolism in women; topical contraceptives seem to have the lowest risk. In predictive models, sex is factored into the evaluation of venous thromboembolism recurrence. The HERDOO2 rule is the only validated rule to help predict the risk for a venous thromboembolism in both sexes. The HERDOO2 rule is named for the following four risk factors used to determine the risk for a venous thromboembolism recurrence: (i) hyperpigmentation, edema, or redness in either leg (1 point); (ii) D-dimer >250 μg/L (1 point); (iii) obesity, BMI ≥30 (1 point); (iv) older age ≥65 (1 point). Women with ≥2 HERDOO points should continue anticoagulants to prevent recurrence where women with ≤1 HERDOO point can discontinue their anticoagulant treatment. In conclusion, women with a first unprovoked venous thromboembolism and a HERDOO score of 0 to 1 have a low risk of recurrent venous thromboembolism, and they can safely discontinue anticoagulants after completing a short-term treatment.

Venous thromboembolism in Behçet’s disease: from guidelines to everyday practice
Zoubida Tazi (Morocco)

Behçet’s disease is a systemic disease that is characterized by a large clinical polymorphism with a high frequency of dermatological manifestations; in addition, there are ocular, rheumatologic, vascular, and neurological manifestations. Vascular involvement was reported in 1946 by Adamantiades, and it is often called vascular Behçet’s disease. It is observed in 20% to 35% of the patients, and it has specific epidemiological and clinical aspects. Indeed, it is mostly observed in young men without thrombotic risk factors or cardiovascular diseases. All vessels, regardless of type (arterial or venous), size, or location may be affected. These attacks are readily associated multifocal vascular manifestations, and they may be accompanied by a fever and an inflammatory biological syndrome. The venous involvement covers 80% to 90% of the vascular attacks, and it is observed in approximatly one-third of patients. Deep vein thrombosis of the lower limbs remains the most frequent manifestation, and it represents 60% to 70% of the venous locations of the disease. However, the involvement of large venous trunks is classic and particularly serious (cava veins, hepatic veins, cerebral sinuses, etc). Arterial involvement is rare (2% to 7% of cases), resulting in a bleak prognosis because arterial lesions are often multifocal, they can affect all territories, and they may manifest with stenosis, aneurysm, or arterial thrombosis. Vascular involvement is the main life-threatening condition that requires a specific, rapid, and aggressive treatment. While the treatment options are under debate, initiation with an anti-inflammatory drug along with an anticoagulant is often considered necessary. Further data is required to determine the best treatment course.

CACTUS study: how to treat the distal deep venous thrombosis?
Isabelle Quere (France)

Currently, it is unknown what place anticoagulants have in the treatment of distal deep venous thrombosis, but the rationale for using anticoagulant is to prevent a lethal pulmonary embolism or a postthrombotic syndrome occurring after a deep vein thrombosis. There are only three clinical studies published about distal deep vein thrombosis, which only involved a small number of patients (n=228). The first study showed that deep vein thrombosis must be treated over 3 months with anticoagulants (warfarin). The second study showed that, after 6 weeks of oral anticoagulant treatment, the risk of a recurrent thromboembolism event is identical to that of patients receiving no anticoagulants, and, after 12 weeks of treatment, the risk of bleeding is higher in the patients treated with oral anticoagulants for a proximal deep vein thrombosis. The third study demonstrated that, in patients with deep muscle venous thrombosis, therapeutic nadroparine and compression therapy did not lead to additional benefits compared with compression therapy alone.

The CACTUS study, a multicenter, double-blind, randomized, placebo-controlled study, was designed to determine the effectiveness of a 6-week course of a therapeutic dose of low-molecular-weight heparin (nadroparine) injections vs placebo (compression stocking) in patients with a first symptomatic isolated distal deep vein thrombosis. The study aims include determining the rates of proximal deep vein thrombosis and symptomatic pulmonary embolisms after the 6 weeks, determining whether the 6-week treatment decreased the frequency of postthrombotic syndrome at 1 year, and monitoring serious or clinically significant bleedings after 6 weeks of treatment and the patients’ quality of life. The rate of proximal thromboembolic events in the case of distal deep vein thrombosis was less frequent than initially believed. In addition, anticoagulant therapy was not superior to simple compression. There were no serious events (pulmonary embolism) in the placebo group, but the risk of bleeding was significantly higher in patients receiving anticoagulants, which is why it is recommended to use only echo-doppler control in patients with symptomatic distal deep vein thrombosis, without any associated anticoagulant therapy. New studies are necessary to assess the utility of prophylactic or intermediary anticoagulant doses in distal deep vein thrombosis patients.

Post-pulmonary embolism dyspnea: from physiopathology to clinical management:
Olivier Sanchez (France)

The 2014 ESC guidelines recommend that systematic screening for detecting chronic thromboembolic pulmonary hypertension be compulsory in patients with severe dyspnea following a pulmonary embolism. Dyspnea typically occurs 12 months after a pulmonary embolism, and it is more severe and frequent due to scintigraphic sequelae of pulmonary embolisms. In addition, effort dyspnea is common during a pulmonary embolism (20% to 40%). The pathophysiological mechanisms are not well known, but it is likely that the persistence perfusion and ventilation inequalities, as well as increasing physiological dead space, play an important role. Effort dyspnea in a pulmonary embolism requires systematic investigations to identify an under lying cardiovascular disease or chronic thromboembolic pulmonary hypertension. When chronic thromboembolic pulmonary hypertension is suspected, echocardiography and pulmonary scintigraphy are recommended, and confirmation can be obtained using right heart catheterization, pulmonary angiography, and an angioscanner. The 2014 ESC guidelines recommend using echocardiography as a first-line noninvasive investigation for diagnosis in case of suspected pulmonary hypertension with effort dyspnea. Right cardiac catheterization is recommended by the 2015 ESCERS guidelines for patients with an echocardiographic probability of intermediate or high pulmonary hypertension. Clinical trials have demonstrated that ventilation perfusion scintigraphy is more sensitive than multidetector angioscanner (CTPA) for detecting chronic thromboembolic pulmonary hypertension. A clinical prediction score has been proposed for chronic thromboembolic pulmonary hypertension after acute pulmonary embolism, but this model requires external validation before being applied in practice. Risk factors for chronic thromboembolic pulmonary hypertension within an acute pulmonary embolism include a personal history of pulmonary embolism, proximal pulmonary embolism, unprovoked pulmonary embolism, and an increase in systolic pulmonary artery pressure >60 mm Hg.